Author Topic: Test Equipment Anonymous (TEA) group therapy thread  (Read 17687571 times)

diskcopy, Vince, Wolfgang and 26 Guests are viewing this topic.

Offline Kosmic

  • Super Contributor
  • ***
  • Posts: 2554
  • Country: ca
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52950 on: March 20, 2020, 02:22:46 pm »
Also while I’m here, argh. Just spent three sodding hours trying to get an STM32 talking to a PCF8574 to no avail. The PCF8574 never acks the first bus command. Bus is properly terminated. Addressing correct. Will attack again tomorrow.

OK lads, I'm taking bets on what BD is doing to dump on ebay to fund buying a logic analyser.

What ??  the siglent can't do the job ?  ^-^
 

Offline bd139

  • Super Contributor
  • ***
  • Posts: 23059
  • Country: gb
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52951 on: March 20, 2020, 02:27:42 pm »
Also while I’m here, argh. Just spent three sodding hours trying to get an STM32 talking to a PCF8574 to no avail. The PCF8574 never acks the first bus command. Bus is properly terminated. Addressing correct. Will attack again tomorrow.

OK lads, I'm taking bets on what BD is doing to dump on ebay to fund buying a logic analyser.


I've got a logic analyser. One of those shonky Saleae clones  :-DD

It's working now. I enumerated the bus with HAL_I2C_Master_Transmit until something acked. It turned up on 0x70 not the expected 0x40 but I don't know why yet. I am flashing an LED off the slave IO port though which is progress!

Currently attempting to persuade the SDS1202X-E to do its protocol decoder thing so I don't have to go in the cupboard and get the Saleae
 

Offline mnementh

  • Super Contributor
  • ***
  • Posts: 17541
  • Country: us
  • *Hiding in the Dwagon-Cave*
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52952 on: March 20, 2020, 02:30:19 pm »


...I've been using SS Brillo in my HAKKO soldering stand for years; it works great and no appreciable wear on my tips, some of which are a decade old.

Good to know :-+

Quote from: mnementh
There's one other great thing about the SS Brillo; SOLDER DOESN'T STICK TO IT like the brass ones. I can wash the stuff with a little Windex and running water in the laundry room sink; then shake it dry and toss it back in the stand. Dump the sink strainer when I'm done, wash my hands and back to my life. I only replace the brillo when it gets ratty and all flyaway strands.

I like the brass wool ones for exactly that reason: the solder sticks to the brass and I can remove the whole contaminated mess with minimal dispersal of lead.

At some point I need to get a better holder than the Hakko 599B, which is a flimsy POC - the upper part doesn't fit onto the base very securely and will come off with the slightest help from the soldering iron >:(

Mine all suffered the same fate... the top split out right on the swaged edge that's supposed to press-fit onto the bottom; it was paper-thin. None of them lasted more than a year; I expect all came from the same "bad run". :-//   I'll admit the single biggest point in the SS scouring pad's favor is the fact they're readily and cheaply available at any grocery store; if you prefer for the solder to stick, no reason you couldn't substitute a copper "Chore-Boy" or similar scouring pad.
:-+

Idiots down here: https://mobile.twitter.com/CrimeLdn/status/1240376132056342540

 :palm: :palm: :palm: :palm:

Best reply:



Fucking wanker.

DidjaEverNotice... when some regressive fuckwit posts video of themselves making a damned fool of themselves... it's invariably 30-60% shaky vid of their feet, and in portrait mode...?

Phones should be made so you CAN'T shoot video unless they're turned sideways...
  :palm:

mnem
alt-codes work here:  alt-0128 = €  alt-156 = £  alt-0216 = Ø  alt-225 = ß  alt-230 = µ  alt-234 = Ω  alt-236 = ∞  alt-248 = °
 
The following users thanked this post: bd139, grizewald

Offline mansaxel

  • Super Contributor
  • ***
  • Posts: 3558
  • Country: se
  • SA0XLR
    • My very static home page
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52953 on: March 20, 2020, 02:31:41 pm »

What I meant was that by "low alcohol" beer most people mean something < 1.0% ABV that you can have a couple and still safely and legally drive. So calling a 3.5% ABV session beer "low alcohol" suggests the writer has a fondness for strong drink. Mansaxel is Swedish, there his neighbours the Finns have the same humorous reputation in regards to drink as the Irish do here in England. In Sweden they call drinking yourself to death "Finnish Suicide".

Yeah, i did use "low" sortakinda tongue in cheek, since we've got all the other sorts of low alcohol beers here too, all the way down to "nothing" which is -- to me -- an incentive to ask for tap water instead. As Grizewald wrote, the 3,5% ABV is the strongest one sold outside the monopoly stores, and done well, it approaches drinkable.  As a matter of fact I'm not that much for the stronger ones, unless the brewing method mandates it. Stout or the strange Belgian variants for instance. I usually buy those for mid-winter consumerism hightides, and have ended up finishing a 75cl N'Ice Chouffe on my own. Sufficient, I'd say, what with it being some 7% ABV or so.

Offline med6753

  • Super Contributor
  • ***
  • Posts: 11320
  • Country: us
  • Tek nut
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52954 on: March 20, 2020, 02:37:29 pm »
Changes to the grocery store I shop at....

Starting Saturday they will now open at 7 AM rather than 6 AM. But on Tuesday, Wednesday, and Thursday they will open at 6 AM until 7 AM for ONLY individuals over 60 or people identified as high risk. Perfect...that's when I'll do my shopping.  :-+

The benefits of being an old grouchy fart.  :-DD
An old gray beard with an attitude.
 
The following users thanked this post: Specmaster

Offline Kosmic

  • Super Contributor
  • ***
  • Posts: 2554
  • Country: ca
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52955 on: March 20, 2020, 02:43:14 pm »
Currently attempting to persuade the SDS1202X-E to do its protocol decoder thing so I don't have to go in the cupboard and get the Saleae

Good luck with that. I tried with my Owon and just abandoned the idea. Saleae software is way easier and better.
 

Offline Kosmic

  • Super Contributor
  • ***
  • Posts: 2554
  • Country: ca
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52956 on: March 20, 2020, 02:51:55 pm »
Well this Covid-19 bastard has certainly left ebay a desert of nothing interesting  :--

Different scenario over here. Just scored a Tektronix RSA3308A Real Time Spectrum Analyzer with a slight problem in the 3.5-8Ghz range  ;D. Will hopefully be an easy fix and will go back on ebay asap.

Oh, you got that one? Cool! I look forward to seeing much more about it.

Yeah, was not planning to bid on it because of the exchange rate and the fact that I need to have it shipped to my PO box across the border (seller did not want to use ebay GSP). But I decided to setup a snipe on gixen just in case. Look like I was lucky :) Hope I din't steal it from you.

With the current situation it's unclear if I will be able to collect it any time soon though  :-\

Crap, the package is still in transit and the US-CAN border is supposed to close today …  |O

Package was delivered, the border close officially tomorrow morning. I should be able to do a quick in and out later today  :-+

God, what I would do for TEA  :palm:
« Last Edit: March 20, 2020, 03:08:06 pm by Kosmic »
 
The following users thanked this post: mnementh

Offline bd139

  • Super Contributor
  • ***
  • Posts: 23059
  • Country: gb
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52957 on: March 20, 2020, 02:52:09 pm »
Currently attempting to persuade the SDS1202X-E to do its protocol decoder thing so I don't have to go in the cupboard and get the Saleae

Good luck with that. I tried with my Owon and just abandoned the idea. Saleae software is way easier and better.

Sussed it. Decode SDA trigger was pointing at the wrong channel. Could be clearer that you need a serial trigger AND the decode trigger thresholds set up separately. They look the same in the menu. That'll teach me not to RTFM

Quick video. Excuse the voice - it's not great today...


 
The following users thanked this post: xrunner, Specmaster, Mortymore, Kosmic

Offline Cerebus

  • Super Contributor
  • ***
  • Posts: 10576
  • Country: gb
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52958 on: March 20, 2020, 03:03:03 pm »
...
In some derivatives of the STM32 family, the I2C peripheral is broken in many interesting ways. Anyway, with STM32 I found bitbanging I2C way much easier than using the I2C peripheral at all.

Thanks for this scary piece of information  ...

The STM32 series are in many respects excellent microcontrollers, in other respects they are a heap of shit. If using an STM32 always, always, obtain a copy of the documented errata for the chip you have in hand. Treat the main documentation (which doesn't get updated with the errata) for the microcontroller as "this is how we hoped it would behave" and the errata as "this is how it does behave, as far as we know, for the moment". The less widely used the subsystem (or the mode the subsystem is used in), the more likely it still has bugs lurking to be discovered.
Anybody got a syringe I can use to squeeze the magic smoke back into this?
 
The following users thanked this post: bitseeker

Offline bd139

  • Super Contributor
  • ***
  • Posts: 23059
  • Country: gb
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52959 on: March 20, 2020, 03:06:03 pm »
I'm learning that slowly. Also ST HAL is a pile of shit. What I put in as the I2C address and what comes out are completely different and I can't see how it got to the destination from the source. Trying to work out where the fault lies and I'm starting to assume it's not in my head.

I suppose the question is: is there a series of microcontrollers with some grunt that isn't a pile of shit? Needs to run FreeRTOS or something as I need multiple tasks.

Edit: I suppose at least the ST Link has half decent debugger built in  :-DD
« Last Edit: March 20, 2020, 03:16:26 pm by bd139 »
 

Offline Cerebus

  • Super Contributor
  • ***
  • Posts: 10576
  • Country: gb
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52960 on: March 20, 2020, 03:12:59 pm »
Typically they organise everything in a hub/spoke model for distribution balancing purposes so they can feed in from one supplier and out to multiple stores. That means total system capacity is defined by the throughput of each spoke, the redundancy model of each spoke and the capacity of the distribution centre to store and route products. If someone places demand then the capacity is entirely limited by the narrowest point and that is always defined to be as cheap as possible on a normal basis, not effective in a crisis, because of the race to the bottom pricing of our supermarkets has squeezed every bit of economy out of the situation.

Fortunately this does seem to be getting better. The stuff in my local East London Sainsbury's that in a sane world ought to be coming from Essex or Kent is coming from Essex or Kent (lettuces and apples are one example I have certain knowledge of).

However, it's not beyond the corporate world to indeed use local produce (because environmentally minded customers have a preference for that) but still move Kentish apples to a Nottingham distribution centre and then back to East London  - what matters is appearances, not facts. I don't think that's happening, but experience with 'Enterprise Systems' means that if it is I would express exactly zero fucking surprise at finding that out.
Anybody got a syringe I can use to squeeze the magic smoke back into this?
 
The following users thanked this post: bd139

Offline drussell

  • Super Contributor
  • ***
  • Posts: 1855
  • Country: ca
  • Hardcore Geek
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52961 on: March 20, 2020, 03:27:52 pm »
Crap, the package is still in transit and the US-CAN border is supposed to close today …  |O

You do know that the border is only closing to non-essential personal travel, right?

The border remains open for shipping goods. 

Both countries need their supply chains affected as little as possible right now...
 

Offline Kosmic

  • Super Contributor
  • ***
  • Posts: 2554
  • Country: ca
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52962 on: March 20, 2020, 03:29:50 pm »
Crap, the package is still in transit and the US-CAN border is supposed to close today …  |O

You do know that the border is only closing to non-essential personal travel, right?

The border remains open for shipping goods. 

Both countries need their supply chains affected as little as possible right now...

I'm pretty sure that crossing the border to get to my PO box (in the US) will be in the non essential category  :(
 

Offline Ice-Tea

  • Super Contributor
  • ***
  • Posts: 3107
  • Country: be
    • Freelance Hardware Engineer
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52963 on: March 20, 2020, 03:34:34 pm »


Dug up a PNP transistor, played around a bit and came to two conclusions:

- I have no idea what I'm doing with this thing
- This thing just works, doesn't it?

« Last Edit: March 20, 2020, 03:37:02 pm by Ice-Tea »
 
The following users thanked this post: Cubdriver, BU508A, Mortymore

Offline med6753

  • Super Contributor
  • ***
  • Posts: 11320
  • Country: us
  • Tek nut
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52964 on: March 20, 2020, 03:43:37 pm »
It's sort of working. That trace should not be tilted like that. You got some issues.
An old gray beard with an attitude.
 

Online AVGresponding

  • Super Contributor
  • ***
  • Posts: 4802
  • Country: england
  • Exploring Rabbit Holes Since The 1970s
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52965 on: March 20, 2020, 03:47:08 pm »
Oh, everywhere has a local brew of "tramp juice". In the UK it's "Special Brew" currently 7.5% ABV, used to be 9%. (Ob Fact: My ex-wife had a peripheral hand in the making of "Crucial Brew" a 9% ABV variant of Red Stripe which came about when someone accidentality fed neat beer (full brew strength as opposed to diluted) to the Charles Wells canning line. The cans thus created were given away to staff, they developed a taste for it, and a new product was born. The name came from people putting on a fake Jamaican accent and describing the brew as "well crucial", which was faddish slang among British youth at the time for "good".)

What I meant was that by "low alcohol" beer most people mean something < 1.0% ABV that you can have a couple and still safely and legally drive. So calling a 3.5% ABV session beer "low alcohol" suggests the writer has a fondness for strong drink. Mansaxel is Swedish, there his neighbours the Finns have the same humorous reputation in regards to drink as the Irish do here in England. In Sweden they call drinking yourself to death "Finnish Suicide".

Oi! Wot a bleedin' liberty...

It used to be 9.8% didn't it?

And it's far too expensive to be "Tramp Juice" (Hobo Juice for the US contingent) at £2 a tinny. Tramp juice is Kestrel Super Strength or Tennents Extra Strength which used to be over 10% ABV. Then there's White Lightning cider, and Buckfast fortified wine...

As for low alcohol beer and driving, I refuse to entertain such sophistry. My policy is no alcohol when driving, period (unless we're talking Gran Turismo or Grand Theft Auto, where it's pretty much mandatory).


Looking forward to the probe tips arriving, will make the 1120A a useful piece of kit rather than a white elephant mouse. Without any 10x or 100x tips, its dynamic range is a mere 500mV.



EDIT:
It's sort of working. That trace should not be tilted like that. You got some issues.

Wow yeah, that's a big rotation on the trace!
« Last Edit: March 20, 2020, 03:49:33 pm by ThickPhilM »
nuqDaq yuch Dapol?
Addiction count: Agilent-AVO-BlackStar-Brymen-Chauvin Arnoux-Fluke-GenRad-Hameg-HP-Keithley-IsoTech-Mastech-Megger-Metrix-Micronta-Racal-RFL-Siglent-Solartron-Tektronix-Thurlby-Time Electronics-TTi-UniT
 
The following users thanked this post: Specmaster

Offline med6753

  • Super Contributor
  • ***
  • Posts: 11320
  • Country: us
  • Tek nut
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52966 on: March 20, 2020, 03:47:42 pm »
Things are gonna get real ugly in these parts.....literally.

Effective Saturday all barbers, nail and hair salons, etc are to close. I pity the men with wives/girlfriends because they are gonna have to listen to a lot of fussing and bitching AND be trapped in the house with them.  :o :-DD

So glad not here.  :phew: 

 
An old gray beard with an attitude.
 
The following users thanked this post: AVGresponding

Offline Kosmic

  • Super Contributor
  • ***
  • Posts: 2554
  • Country: ca
 

Offline McBryce

  • Super Contributor
  • ***
  • Posts: 2705
  • Country: de
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52968 on: March 20, 2020, 03:50:01 pm »
That's definitely not what you should be seeing on the screen. It may have an issue or you may just have some settings way off. Test it with a Diode first as the diode curve is much easier to evaluate.

McBryce.
30 Years making cars more difficult to repair.
 

Offline Cerebus

  • Super Contributor
  • ***
  • Posts: 10576
  • Country: gb
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52969 on: March 20, 2020, 03:53:32 pm »
I'm learning that slowly. Also ST HAL is a pile of shit. What I put in as the I2C address and what comes out are completely different and I can't see how it got to the destination from the source. Trying to work out where the fault lies and I'm starting to assume it's not in my head.

I suppose the question is: is there a series of microcontrollers with some grunt that isn't a pile of shit? Needs to run FreeRTOS or something as I need multiple tasks.

Edit: I suppose at least the ST Link has half decent debugger built in  :-DD

I'm currently playing with chibios/giving it a try. It has its own HAL.

Points in its favour:
  • Appears to be written by someone with half a clue*,
  • Supports a wide range of chips with source code compatibility from, say, an MSP430 to an STM32,
  • Good architectural documentation,
  • supports a serial "printf" style debug port for old school eschewers, like me, of fancy debuggers ("If you can't debug it with printf and vi you're NFG"),
  • Eclipse and command line build and debug support (again, "give me make or give me death" - I really hate having to learn how to beat the IDE into submission before I can get even "hello world" done).

Points against:
  • lousy or non-existent application level programmer documentation.


*The supplied 'blinky' for these also runs the test suite for the RTOS with results output over the serial debug port.
Anybody got a syringe I can use to squeeze the magic smoke back into this?
 
The following users thanked this post: bitseeker, bd139

Offline med6753

  • Super Contributor
  • ***
  • Posts: 11320
  • Country: us
  • Tek nut
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52970 on: March 20, 2020, 04:01:37 pm »
The shoe dropped....New York State is going on full lockdown Sunday night.  :scared: :scared:
An old gray beard with an attitude.
 

Offline Mortymore

  • Frequent Contributor
  • **
  • Posts: 467
  • Country: pt
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52971 on: March 20, 2020, 04:10:28 pm »

Dug up a PNP transistor, played around a bit and came to two conclusions:

- I have no idea what I'm doing with this thing
- This thing just works, doesn't it?

It looks like it is vertically inverted, and at least horizontally over sized. It seems like the curve tracing start point is on the upper left corner, when it should be on the lower left corner.
And like med6753 said, it looks a bit tilted.

This is how a curve trace for an NPN BC547B looks like in my Tek2445

 
The following users thanked this post: edavid, BU508A

Offline bd139

  • Super Contributor
  • ***
  • Posts: 23059
  • Country: gb
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52972 on: March 20, 2020, 04:13:02 pm »
I'm learning that slowly. Also ST HAL is a pile of shit. What I put in as the I2C address and what comes out are completely different and I can't see how it got to the destination from the source. Trying to work out where the fault lies and I'm starting to assume it's not in my head.

I suppose the question is: is there a series of microcontrollers with some grunt that isn't a pile of shit? Needs to run FreeRTOS or something as I need multiple tasks.

Edit: I suppose at least the ST Link has half decent debugger built in  :-DD

I'm currently playing with chibios/giving it a try. It has its own HAL.

Points in its favour:
  • Appears to be written by someone with half a clue*,
  • Supports a wide range of chips with source code compatibility from, say, an MSP430 to an STM32,
  • Good architectural documentation,
  • supports a serial "printf" style debug port for old school eschewers, like me, of fancy debuggers ("If you can't debug it with printf and vi you're NFG"),
  • Eclipse and command line build and debug support (again, "give me make or give me death" - I really hate having to learn how to beat the IDE into submission before I can get even "hello world" done).

Points against:
  • lousy or non-existent application level programmer documentation.


*The supplied 'blinky' for these also runs the test suite for the RTOS with results output over the serial debug port.

Yes ChibiOS is on the list of things to evaluate for sure. Let us know how you get on with it. My history is getting FreeRTOS working on AVR. Before that was VxWorks on PPC cores inside Xilinx stuff about 20 years ago so I am rusty as hell :)

I sussed the problem out and now it makes sense. Two missing bits of information through two bits of idiocy on my part:

1. The IC I have is a PCF8574A not a PC8574 which has a different I2C addressing scheme. So I should be farting stuff at 0x38.
2. The HAL_I2C_Master_Transmit documentation [1] says that DevAddress needs to be shifted left one bit, I assume because it just sets the write bit and slings it out the IO port.

This is the world's most complicated LED flasher :)

Code: [Select]
    uint8_t b_on = 0xff;
  uint8_t b_off = 0x00;
  uint8_t addr = 0x38; // PCF8574 A0=0,A1=0,A2=0
    while (1)
    {
    /* USER CODE END WHILE */
    HAL_I2C_Master_Transmit(&hi2c1, addr << 1, &b_off, 1, 100);
    HAL_Delay(500);
    HAL_I2C_Master_Transmit(&hi2c1, addr << 1, &b_on, 1, 100);
    HAL_Delay(500);
    }

Any day you learn something however small or however large is a good day. Also importantly working out why something is working and not that it just does work is a good day.

Anyway back to less interesting things ... GDPR training!  >:(

[1] http://www.disca.upv.es/aperles/arm_cortex_m3/llibre/st/STM32F439xx_User_Manual/group__i2c__exported__functions__group2.html#ga9440a306e25c7bd038cfa8619ec9a830
« Last Edit: March 20, 2020, 04:15:40 pm by bd139 »
 

Offline BU508A

  • Super Contributor
  • ***
  • Posts: 4535
  • Country: de
  • Per aspera ad astra
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52973 on: March 20, 2020, 04:38:48 pm »

Dug up a PNP transistor, played around a bit and came to two conclusions:

- I have no idea what I'm doing with this thing
- This thing just works, doesn't it?

It looks like it is vertically inverted, and at least horizontally over sized. It seems like the curve tracing start point is on the upper left corner, when it should be on the lower left corner.
And like med6753 said, it looks a bit tilted.

This is how a curve trace for an NPN BC547B looks like in my Tek2445



Just for comparison: could you do a PNP, e.g. a BC 557 or so?

Thanks
“Chaos is found in greatest abundance wherever order is being sought. It always defeats order, because it is better organized.”            - Terry Pratchett -
 

Offline capt bullshot

  • Super Contributor
  • ***
  • Posts: 3033
  • Country: de
    • Mostly useless stuff, but nice to have: wunderkis.de
Re: Test Equipment Anonymous (TEA) group therapy thread
« Reply #52974 on: March 20, 2020, 04:39:43 pm »
I'm learning that slowly. Also ST HAL is a pile of shit.

Ack. I use HAL only to route / initialize the pins, clock system and some other basics. Often basic initialization of a peripheral through HAL does the job good enough so I don't need to cope with all the details. For the real stuff (anything after basic initialization) I access the HW directly where it is needed or write my own basic driver layer.

Quote
I suppose the question is: is there a series of microcontrollers with some grunt that isn't a pile of shit? Needs to run FreeRTOS or something as I need multiple tasks.
There are many series of microcontrollers with some grunt available, but you'll find you pile of shit with them too, maybe at some other place.
HW wise the STM32 aren't too bad, for some peripherals the others are better, for some other peripherals STM32 is better. CortexM and it's typical associated toolchains is all over the place anyway and no need to switch to another architecture.

Quote
Edit: I suppose at least the ST Link has half decent debugger built in  :-DD
Works for me. The ST Link debugger exists long enough to evolute from absolute crap to a decent useful level, the modern ones also incorporate a working USB CDC for printf() debugging onboard (nearly all nucleo boards have that).
Safety devices hinder evolution
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf